Conservation detection dog handler teams (CDDHTs) offer many potential benefits to the world of conservation. Seabird populations are an important component of marine ecosystems. However, they are threatened by several anthropogenic activities, including the introduction of invasive species. Although CDDHT can support seabird conservation through invasive species management efforts and population assessments, they are under-utilised. A lack of methodological standardisation within CDDHT work and the under-publishing of their use within seabird research leads to difficulties in conducting new CDDHT seabird-related studies due to an inability to learn from previous research. This study aimed to address these shortcomings by investigating the techniques and methods used by those actively working with, or planning to work with, CDDHT on a seabird project to better understand them, and propose best practices in the field. Seven professionals who have used, or will use, CDDHT as part of a seabird project (four handlers, three ecologists/researchers) participated in structured written surveys which were thematically analysed. Five superordinate themes emerged from the survey data: Training, Location, Role of Handler, Wildlife Considerations, and Dog Selection Criteria, with the first two themes having several subordinate themes. A summary of best practices was developed from the findings, with notable recommendations including preparation across all project elements, networking with other professionals, and making judgments on the use of techniques like discrimination and field trials based on the specific project and dog(s). These results can serve to benefit future seabird studies involving CDDHT as well as supporting the development of standardisation in the CDDHT field.

Conservation detection dogs (CDDs) are defined as working dogs Canis familiaris that use their exceptional olfactory abilities (Kokocińska-Kusiak et al. 2021) to support conservation projects (MacKay et al. 2008; Helton 2009; Woollett et al. 2013) by operating alongside a human handler to create a conservation detection dog handler team (CDDHT; Richards et al. 2021). CDDHTs can be complementary to current animal monitoring techniques, such as capture-mark-recapture, camera trapping, playback surveys and other surveys. Indeed, CDDHTs can cover large distances in short timespans and find many samples, often greater numbers than a human observer working alone might encounter (Browne et al. 2006; MacKay et al. 2008; Kerley 2010; Grimm-Seyfarth & Klenke 2018; Stanhope & Sloan 2019). CDDHTs are also unaffected by sampling bias arising from use of visual information (MacKay et al. 2008; Kerley 2010), which can also be invasive (Browne et al. 2006; Kerley 2010; Grimm-Seyfarth & Klenke 2018; Richards 2018). However, CDDHTs are costly in time and money, as it can take months or years to train a CDDHT and the CDD must be maintained through food, housing, transport and training (MacKay et al. 2008; Kerley 2010). Additionally, acquiring appropriate training samples can be practically and legally challenging, depending on the target species (MacKay et al. 2008; Kerley 2010). Despite these limitations, CDD methodology has been used worldwide (Grimm-Seyfarth et al. 2021) for more than 100 years (Hill & Hill 1987), detecting over 400 animal species including 114 bird species, and nine seabirds (Grimm-Seyfarth et al. 2021).

Seabirds are an important component of both marine and terrestrial ecosystems (Schreiber & Burger 2001). On one hand, due to their dependence on the marine environment for trophic resources and their high ranking in the marine food web, seabirds are considered good bioindicators of the health and functioning of the marine environment. On the other hand, because they breed on land, their guano, food remains and other components (egg remains, carcasses, etc.) improve the fertility of marine and terrestrial areas around their colonies (Croxall et al. 2012; Paleczny et al. 2015; Dias et al. 2019; Rodríguez et al. 2019). Due to their biology, they are exposed to threats both on land and at sea and therefore, they are also one of the most at-risk groups of birds globally, with 70% of seabird populations suffering declines (Grémillet & Boulinier 2009; Paleczny et al. 2015; Rodríguez et al. 2019). Threats to seabirds include invasive alien species like Rats Rattus sp. and Cats Felis catus, fisheries bycatch, overfishing, climate change, hunting and trapping, disturbance, problematic native species, energy production and mining and building developments (Spatz et al. 2017; Dias et al. 2019). Population estimates play a key role in understanding causes of their decline as well as their conservation management (Paleczny et al. 2015). One of the most basic aspects of obtaining seabird population estimates involves finding breeding locations and censusing breeding pairs. However, many seabird species employ cryptic behaviours to limit predation and piracy, such as nesting on remote islands or in inaccessible areas, such as within burrows and crevices or on cliffs (Schreiber & Burger 2001), making them difficult for human researchers to detect and monitor.

To produce and maintain an operational CDDHT to a high standard, both the dog and handler require extensive training. There are numerous ways to train a detection dog, but generally the process consists of search development, training a passive indication, imprinting on the target odour and any necessary discrimination training. Discrimination training is undertaken to ensure a CDD can correctly discriminate trained odours from non-trained odours (Porritt et al. 2015). Non-trained odours can include training aids such as nitrile gloves and storage jars, as well as other odours the dog may experience in the field. This may include scents that are similar to the target odour, for example, a nontarget species of seabird.

How long training a detection dog takes is variable, and dogs also require continuation training throughout their working careers. The length of time it takes to train the handler will also vary depending on the ability of the selected individual to recognise changes in behaviour and indication of their detection dog (DeMatteo et al. 2019). The pairing of dog handler teams is also important, and time must be given for new teams to bond as this can impact detection performance (Jamieson 2018).

The time taken to search a given area depends on a variety of factors including terrain, temperature, humidity, wind direction and speed. The location of the target odour can impact how intensely the dog needs to search, potentially increasing search time, e.g. if the target odour is buried (Osterkamp 2020).

CDDHTs can help overcome difficulties in conducting seabird population surveys due to their lack of reliance on visual information and ability to cover large areas more quickly than humans (Kerley 2010). In the published literature, however, there are limited examples of CDDHTs being deployed to assist seabird conservation efforts. CDDHTs are most commonly used as part of biosecurity measures to avoid the introduction of invasive species and for invasive species detection during and following eradication efforts (Russell et al. 2008; Bellingham et al. 2010; Pierce et al. 2015; Robinson et al. 2015; Springer 2018; Phillips 2019; Robinson & Gadd 2020). In the few cases where CDDHTs have been actively used to detect seabirds, they have found nests or burrows of species like Little Penguin Eudyptula minor (Cargill et al. 2022), Band-rumped Storm-petrel Hydrobates castro (Galase 2019), Black Petrel Procellaria parkinsoni (Bell et al. 2014), and Manx Shearwater Puffinus puffinus (Bolton et al. 2021). Although CDDHTs can be costly to hire, it is important to understand why they have been under-utilised in seabird conservation management given their potential benefits.

Perhaps a critical concern in utilising CDDHTs for seabird conservation is the lack of methodological consistency observed across studies. This inconsistency hampers the ability to evaluate the effectiveness of this approach and emphasises the need for its standardisation (Bennett 2015; Johnen et al. 2017; Hayes et al. 2018; Otto et al. 2019). Persistent problems that occur in CDD studies include a lack of methodological detail on training and searches (Johnen et al. 2017; Bennett 2020), small sample sizes when assessing CDDHT performance (Lazarowski et al. 2020; Whitehouse-Tedd et al. 2021), and inconsistency regarding the measures used to evaluate a CDD9s performance (Johnen et al. 2017; Hayes et al. 2018). Furthermore, it appears that much of the literature regarding practical seabird conservation, both including and excluding CDDHTs, goes unpublished, as exampled by the following studies which refer to and use findings and data from unpublished work: Russell et al. 2008; Bellingham et al. 2010; Bell et al. 2014; Parker & Rexer-Huber 2015, 2016; Robinson et al. 2015; VanderWerf & Young 2017; Cargill et al. 2022. This means that researchers are prevented from learning from previous studies and may be making the same mistakes as others, thus exacerbating the need to move towards standardisation of survey methods used.

Although there are literature reviews and books that outline key aspects of using working dogs including CDD (MacKay et al. 2008; Helton 2009; Beebe et al. 2016; Lazarowski et al. 2020), we were unable to find any examples of thematic analyses or qualitative assessments of methodologies within the CDD literature. Gathering insights from professionals who work with CDDs for seabird projects would allow for the assessment of methods and techniques that are currently used in the field as well as the rationale for their use. Furthermore, consolidating this information could help researchers, environmental non-governmental organisations, statutory conservation bodies or developers considering using a CDDHT in a seabird project to make decisions regarding selecting a qualified team, and improve their understanding and support of the CDDHT method. In this study, structured written interviews were conducted with professionals who have used, or plan to use, CDDHTs in seabird conservation. Our aim was to explore different aspects of project design and the utilisation of CDDs, with the goal of developing best practice recommendations.

The authors acknowledge the input and support that was provided by all participants in this study. Authors would like to thank Nicole Galase from the Center for Environmental Management of Military Lands (CEMML) and Teresa Gajate, dog handler for CEMML for their input at the beginning of this project. The work in Northern Ireland was supported through the National Parks and Wildlife Service (NPWS) of the Republic of Ireland, and the Marine Protected Area Management and Monitoring (MarPAMM) project which is funded by the European Union's INTERREG VA Programme as managed by the Special EU Programmes Body (SEUPB), and includes the Agri-Food and Biosciences Institute (AFBI), Marine Scotland, Scottish Natural Heritage, University College Cork, Ulster University, Scottish Association for Marine Science, and BirdWatch Ireland as project partners. We thank the staff in this project for their support, including but not limited to Kendrew Colhoun and Naomi Wilson of Agri-Food and Biosciences Institute (AFBI).

Balzani, A. & Hanlon, A. 2020. Factors that influence farmers' views on farm animal welfare: a semi-systematic review and thematic analysis. Animals 10: 1524. https://doi.org/10.3390/ani10091524

Beebe, S. C., Howell, T. J. & Bennett, P. C. 2016. Using scent detection dogs in conservation settings: a review of scientific literature regarding their selection. Frontiers in Veterinary Science 3: 96. https://doi.org/10.3389/fvets.2016.00096

Bell, E. A., Mischler, C., Sim, J. L., Scofield, P., Francis, C., Abrahams, E., & Landers, T. 2014. At-sea distribution and population parameters of the black petrels (Procellaria parkinsoni) on Great Barrier Island (Aotea Island), 2013/2014. Report to the Conservation Services Programme, Department of Conservation, Wellington, New Zealand.

Bellingham, P. J., Towns, D. R., Cameron, E. K., Davis, J. J., Wardle, D. A., Wilmshurst, J. M., Mulder, C. P. H. 2010. New Zealand island restoration: seabirds, predators, and the importance of history. New Zealand Journal of Ecology <.i>34: 115–136.